From: Xeonacid <h.dwwwwww@gmail.com>
Date: Tue, 9 Jun 2026 14:03:02 +0200
Subject: Avoid hard-coding algorithms doctest

Stemmer.algorithms() returns the list from the linked libstemmer
library, so the exact set can change as libstemmer adds or removes
algorithms. A quickstart doctest that hard-codes the whole list is
therefore brittle even though the API behavior is still correct.

Use doctest ellipsis to allow the list to grow without causing the
test to fail.  This keeps the example executable and still checks the
shape of the output without tying the documentation to one exact
libstemmer release.

Fixes: #61

Co-authored-by: Olly Betts <olly@survex.com>
---
 docs/quickstart.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/quickstart.txt b/docs/quickstart.txt
index 73235e3..ed006ab 100644
--- a/docs/quickstart.txt
+++ b/docs/quickstart.txt
@@ -9,8 +9,8 @@ First, import the library:
 
 Just for show, we'll display a list of the available stemming algorithms:
 
->>> print(Stemmer.algorithms())
-['arabic', 'armenian', 'basque', 'catalan', 'danish', 'dutch', 'dutch_porter', 'english', 'esperanto', 'estonian', 'finnish', 'french', 'german', 'greek', 'hindi', 'hungarian', 'indonesian', 'irish', 'italian', 'lithuanian', 'nepali', 'norwegian', 'porter', 'portuguese', 'romanian', 'russian', 'serbian', 'spanish', 'swedish', 'tamil', 'turkish', 'yiddish']
+>>> print(Stemmer.algorithms())  # doctest: +ELLIPSIS
+[...'arabic', ...'english', ...'yiddish'...]
 
 Now, we'll get an instance of the english stemming algorithm:
 
